Dohoon (Don) Lee

Seoul, Korea

I received my B.S. in Electrical and Electronic Engineering from Yonsei University and completed my M.S. in Artificial Intelligence at Seoul National University (SNU IPAI)

I am currently focused on research at the intersection of generative modeling, trajectory optimization, and reinforcement learning, with a broader interest in building intelligent systems capable of learning and planning in complex environments.

My recent work explores the theoretical and practical aspects of flow- and diffusion-based models, including simulation-based optimization and sample-adaptive inference strategies.

I am actively seeking Ph.D. opportunities and research internships starting in 2026, and I’m particularly interested in collaborating with groups working on fundamental advances in machine learning, world modeling, and control.

news

May 01, 2025	Our paper, “Multidimensional Adaptive Coefficient for Inference Trajectory Optimization in Flow and Diffusion,” has been accepted to ICML 2025!
Dec 09, 2024	Began my mandatory military service at the Army AI Center, Headquarters.

selected publications

Multidimensional Adaptive Coefficient for Inference Trajectory Optimization in Flow and Diffusion

Dohoon Lee, Jaehyun Park, Hyunwoo J. Kim, and 1 more author

2025

Bib HTML PDF

@misc{lee2025multidimensionaladaptivecoefficientinference,
  title = {Multidimensional Adaptive Coefficient for Inference Trajectory Optimization in Flow and Diffusion},
  author = {Lee, Dohoon and Park, Jaehyun and Kim, Hyunwoo J. and Lee, Kyogu},
  year = {2025},
  eprint = {2404.14161},
  archiveprefix = {arXiv},
  primaryclass = {cs.LG},
  url = {https://arxiv.org/abs/2404.14161},
}